017 - Results of prophylactic inoculation against pneumococcus in 12,519 men

1918 seems to be an interesting year for vaccine research. Vaccine therapy was on its ebb, and development of serious prophylactic vaccines was on the rise. At least that's how it seems from the number of publications relating to each subject. 

The topic of today's post is a rudimentary clinical trial of a vaccine against pneumococcus¹, now known as Streptococcus pneumoniae, a bacterium that is one of the pathogens that cause pneumonia. Other causes can be viral, fungal, or bacterial (from other species), but at the time, certain bacterial pneumonias had been the cause of up to 80% of the pneumonial deaths in US military training camps in winter.

Previous studies had tried to test immunization against pneumococcus, but there are several different strains (types I,II,III, and IV), and vaccination against one doesn't necessarily protect against the others, and these studies did not discriminate. Type IV is especially difficult because it has a number of sub-varieties, according to this article.

This study¹ focused on one particular military camp, Camp Upton, at which 70% of their pneumonias were pneumococcal (the other 30% were other kinds of Streptococcus). Before the experiment or any vaccinations, cases of pneumonia had been spread fairly evenly through the groups that would later make up the vaccinated and unvaccinated groups, and 1/3 of those cases were types I to III, another 1/3 was type IV, and the last 1/3 was other bacteria.

The preparation of vaccine was similar to previous studies: organisms (pneumococci of types I, II, and III) were grown in broth culture, killed by heating, and the resulting vaccine preserved with 0.3% tricresol. Those that worry about toxins might be wary of tricresol, which apparently is harmful in high concentrations but not in low, at least not acutely; not much is known about long-term effects of low levels of exposure. But that's not super-relevant in this case. 

The authors determined the optimum total dose and dosing intervals using 42 volunteers, testing the antibody levels of their serum (via agglutination and protection of mice against the live bacteria), and concluded that the 6-9 billion cells was the best dose, and spreading it over four separate injections (about a week apart) reduced the severity of negative side effects. For these volunteers, at least, there was hardly any negative reaction to this inoculation schedule.

Using this procedure, the authors then tested the vaccine on 12,519 of the soldiers at Camp Upton, NY (about 40% of the men). These were seasoned troops in stable positions at the camp. Out of that number, 25 of them (1 in 500) had an adverse reaction serious enough that they spent at least a short time away from their duties. Some of these 25 only received one or two of the scheduled 4 doses, due to the severity of their reactions. Those that did react at all complained of malaise, chills, fever, and muscle pains; some had something like a respiratory infection; others reported that infections they had previously were worse.

One reaction the authors noted especially was, 152 individuals (1 in 82 men) had strange reactions at the site of injection, with tenderness and infiltration of fluid. It seemed to go away on its own before too long though, but the authors determined that it was due to hypersensitivity to a bacterial toxin.

So finally we get to the effectiveness results. The authors were careful, as much as possible, to determine the true identity of the organism causing pneumonia in each case, which is good.

As mentioned, there were 12519 vaccinated men, and these were compared to about 19481 unvaccinated troops (the number fluctuated a bit due to soldiers coming to and going away from the camp). About 75% (or 14610) of the unvaccinated group were seasoned, stable troops that were probably less susceptible to pneumococcal infection, whereas the other 25% (4871) were new recruits, presumably more susceptible.

These groups were followed for about 10 weeks, from February to April. I made some graphs to represent the results, since it seems like graphs were illegal to publish back then (or probably just difficult to make for a printing press):

Figure 1. Cases per 10,000 men. “Unvax-resistant” group is seasoned men, 75% of total unvaccinated; “Unvax-susceptible” is new recruits; “Unvaccinated” is both combined. “Pneumo I-III” refers to pneumococcus types I to III, against which the vaccinated group was vaccinated. “Pneumo IV” is type IV, not vaccinated against. “Streptococcus” is other streptococcal pneumonias, labeled as “S. haemolyticus” and “S. viridans.” “Other” is pneumonias caused by other agents, including something called “B. influenzae.”

Figure 2. Deaths per 10,000 men. Same labeling as figure 1.

It’s worth noting that there was one case of types I-III in the vaccinated group (you can see the small bar in figure 1), but this case appeared 24 hours after the first dose, so it’s likely that guy was already infected and there wasn’t time for the immunity to kick in yet. So it was effectively zero cases in the vaccinated group, vs. 26 cases in the unvaccinated (about 13 per 10,000 men, with 3 in 10,000 dying from types I-III).

However, a better comparison is between vaccinated and the seasoned unvaccinated group, so it’s more like zero vs. 18 cases (12 per 10,000). Not much different, though I don’t know how many deaths were from the more susceptible group.

Another important question is, how long would this protection last? The study only went for about 2 months, so at least that long, but one can’t really extrapolate from that.

Lastly, it’s curious that the vaccinated group seemed to be significantly protected against things they weren’t vaccinated against, such as the other types of pneumococcus and Streptococcus. Most likely, either the vaccine conferred some cross-protection against other bacteria (possible), or there was some confounding factor that differed between the vaccinated and unvaccinated groups (also possible). Overall, I rate the quality of this evidence as medium.

A number of other publications have cited this one, for various reasons. Of those that commented on its clinical trial aspect, many described its results favorably, though some had some criticisms, such as the following:

“Their experiment is open to obvious criticisms, but the results at least indicate a degree of temporary immunity against the pneumococcal types used in the vaccine.”²

“The validity of the data is lessened by differences in the composition of the vaccinated and control groups, since the vaccinated group was made up entirely of seasoned troops, whereas in the control group approximately 25% were new recruits.”³

“Interpretation of the results was clouded by such variables as differences in the composition of the immunized and control groups; uncertainty as to whether the specific pneumococcal types included in the immunizing preparation were the same as those currently causing pneumonia; failure to determine whether the observed decline in cases in the immunized group was due to a decrease in cases caused by pneumococcal types included in the vaccine; and inadequate control of the antigenicity of the preparations used.”⁴

 “However, since the incidence of other types of pneumonia—particularly those due to the streptococcus—was likewise greater in the controls than in the vaccinated, the validity of this experiment is open to question.”⁵

“Both trials [this and a similar], each including more than 10,000 recipients of vaccine, were suggestive of the prophylactic value of vaccination in reducing the incidence of pneumococcal pneumonia. Careful analysis of the data, however, failed to establish unequivocally such a conclusion.”⁶

The implication, I think, is that better studies have been done after this one; I look forward to reading them.

1. Cecil, R. L. & Austin, J. H. Results of prophylactic inoculation against pneumococcus in 12,519 men. J. Exp. Med. 28, 19–41 (1918).

2. Cruickshank, R. Pneumococcal Infections. The Lancet 221, 680–685 (1933).

3. MacLeod, C. M., Hodges, R. G., Heidelberger, M. & Bernhard, W. G. Prevention of Pneumococcal Pneumonia by Immunization with Specific Capsular Polysaccharides. J. Exp. Med. 82, 445–465 (1945).

4. B., W. H. Immunization with pneumococcus polysaccharide. Ann. Intern. Med. 24, 928–930 (1946).

5. Edsall, G. Active Immunization. N. Engl. J. Med. 241, 60–70 (1949).

6. Austrian, R. The pneumococcus and some men who came to Yale: the Dorothy M. Horstmann Lecture. Yale J. Biol. Med. 66, 315–324 (1993).

016 - Combined Vaccination with Multiple Vaccines (Quadruple, Quintuple, and Sextuple)

One thing I hear from people who question the safety of vaccines is that too many at once may be harmful. This seems implausible, given the vastly larger number of antigens (material recognized as foreign by the immune system) that a person is exposed to in daily life every day.

It seems that even in the early 20th century, people (with reason) questioned the effectiveness of combining antigens from multiple into a single injection. So this study tested the antibody response and adverse reactions of different combinations of bacterial pathogens.

The study tested tetra-valent, penta-valent, and hexa-valent (that is, containing antigens from 4, 5, or 6 bacteria) vaccines in rabbits and people. Typhoid and paratyphoid A and B were present in all combinations, but some also had Vibrio cholerae (which causes cholera), Brucella (which causes Malta fever or brucellosis), or Yersinia pestis (which causes plague).

With all of them, they found that the levels and pattern of antibody responses were similar to those seen when vaccines were given containing only one of the organisms at a time, so there was no deficiency of antibody response.

Regarding adverse reactions, all of them seemed to cause local reactions (redness, inflammation) and general reactions (fever, malaise, pains) for 1-2 days, but the combos containing plague were much more severe. They noted that the combo vaccines with plague actually caused less severe reactions than plague vaccine by itself though. But of those without plague, more than 120,000 people (military folk) received some combinations without any more serious effects. So the authors conclude that the combination vaccines are safe and effective, and much more convenient than monovaccines, especially in tropical military settings.

Citation: Castellani, A. & Taylor, F. Combined Vaccination with Multiple Vaccines (Quadruple, Quintuple, and Sextuple). Br Med J 2, 356–359 (1917).

015 - The duration of immune bodies in the blood after antityphoid inoculation

An important question in the effectiveness of vaccines is how long immunity from them lasts. Depending on the stimulus, the immune system may remember and recognize a pathogen pretty much forever after it is exposed once, but sometimes immunity wanes or even disappears completely, such that an individual can be reinfected with the same pathogen.

The following study is on the antityphoid fever vaccine of the day, how long antibodies from it last after a person is vaccinated. A.E. Wright had previously reported observing people retain some immunity even after two years, and another had claimed to see some immunity after four or even six years in some people.

So the author in this study set out to do a more systematic study, using 24 subjects. Antibodies were measured a) by agglutination, that is, the ability to make bacteria clump together by binding to them, and b) by bactericidal activity.

The results were not great: antibody levels were highest one month after inoculation, but fell off such that after a year, about 80% of the subjects showed no more antibody activity of either type. So the author recommended yearly re-immunization for those at risk for typhoid. Useful if one is in a high-risk area, but not something to give to the general public in a generally typhoid-free area. Good to know though.

Citation: Wollstein, M. The duration of immune bodies in the blood after antityphoid inoculation. J Exp Med 16, 315–324 (1912).

014 - Three Cases of Probable Cancer Transplantation from the Use of the Gilman Vaccine

I skipped 1908 through 1910 because it was pretty much entirely just a big debate about the effectiveness of vaccine therapy, which I've already talked about enough, and since it isn't really used much today (if at all), it isn't in line with my goal of evaluating the safety and effectiveness of modern vaccines.

However, some such studies can be interesting and unique enough to be worth reading and writing about. This is one of them: a bit horrifying, actually.

While most people were trying to use vaccines and vaccine therapy for all kinds of different bacterial infections, some physicians in the Philippines, Gilman and Coca, had the idea of treating cancer with the same method: after removing a tumor, inactivating it somehow and injecting it back into the patient to stimulate an immune response against the material.

Unfortunately, the inactivation procedure apparently didn't always work, so the current series of case reports details several cancer patients who apparently experienced a sort of artificial metastasis when the tumor material injected into them started growing new tumors at the injection site.

The treatment itself sounded rather unpleasant: injection into the abdomen, sometimes with seizures as a side effect. But apparently the method used to inactivate the cancerous material before injection (5% phenol) was inadequate. Fortunately, it seemed like doctors received this article as a warning and were more careful in the future.

On the face of it, the treatment doesn't seem implausible. Cancer cells are different in some ways from healthy tissue, theoretically susceptible to immune attack, so inducing a stronger response against them might be helpful. People even now are trying to induce immune responses against certain cancers (see here). But it is very tricky, considering how similar cancer cells can be to healthy cells (they share the same ancestor cells, after all). It might be possible to cause some autoimmune disorders using this cancer vaccine therapy method, I would think, so it's probably best that it is no longer in use.

Citation: Coffey, W. B. Three Cases of Probable Cancer Transplantation from the Use of the Gilman Vaccine. Cal State J Med 9, 129–130 (1911).

013 - The Opsonic Index as a Guide to regulate the use of Vaccines in the Treatment of Disease

This paper seems to be near to the inception of immunology, obviously an important area of research into the mechanism and function of vaccines. The authors report on work done by a number of researchers (mostly women, interestingly) evaluating the opsonic index, mentioned here before.

To recap: the opsonic or phagocytic index is a measure of the activity of white blood cells in consuming bacteria they encounter.

Research previous to this paper had discovered that the increased activity of the white blood cells (aka leukocytes) of vaccinated animals was not something inherent to the cells themselves—isolated leukocytes from unvaccinated animals had the same activity. Rather, the increased activity came from some other component of the serum of the animal (the liquid, cell-free portion of the blood). A.E. Wright referred to this serum element as "opsonins," which bind to the pathogen and induce phagocytosis, that is, uptake by white blood cells.

According to Wright, who developed the opsonic index method, determining this index value is important for administering vaccine therapy, because too high a dose can actually reduce a person's immunity to a pathogen.

The way this opsonic index is determined is to take equal amounts of bacteria, blood cells, and serum and mix them together, incubate for 15 minutes, then examine them under a microscope and count how many bacteria each leukocyte has taken up into itself.

The authors here describe a number of factors that must be controlled to get an accurate opsonic index value; failure to control them can introduce significant artifical variation. Such factors include number of leukocytes counted from each sample (at least 50 is important, up to 150; speaking from experience, that sounds like a seriously unpleasant experience for the person doing the counting, staring into a microscope for so long; especially since it is also important for just one person to do all the counting, since different people count the numbers differently and induce more variation), and slight variations in mixture proportions can introduce variation too. Because of these difficulties, it often took so long to finish a counting that the vaccinator had to go ahead and treat the patient before actually getting the results.

There was some investigation into the nature of these opsonins, too. Those from unvaccinated serum were more vulnerable to disruption by heating than those from vaccinated, and the former seemed to bind more generally—they could be titrated or removed from the serum using common "complement absorber" compounds—whereas the elements from vaccinated serum were more heat-stable and only bound well to the pathogens that had been vaccinated against. The latter, they referred to as "antibodies." We know more about these elements in modern times, of course, as members of the innate vs. adaptive immune systems.

The authors also report a number of trials of vaccine therapy against a variety of pathogens, using opsonic index measurements to inform the therapy. In most of the cases it didn't seem to help very much, and the index readings were inconsistent, so it didn't seem as useful as Wright made it out to be.

Citation: Park, W. H. & Biggs, H. M. The Opsonic Index as a Guide to regulate the use of Vaccines in the Treatment of Disease. J Med Res 17, 77–88 (1907).

012 - A note on Haffkine's antiplague vaccine

After the last post, I am wary of people trying to use killed bacterial cultures as vaccines. That's probably somewhat unreasonable, as it does seem to work sometimes.

Today's attempt in this area is an anti-plague vaccine, intended to protect against Yersinia pestis, aka the Black Death or bubonic plague. There were some epidemics going on in India in the early 1900s, and this vaccine apparently did not work well as a treatment for plague, but the author wanted to test its effectiveness as a preventative measure.

In the village of Kirkee at the time, there were two somewhat separate populations: 5,640 civilians and 1,312 people and their families employed by the military. These lived in separate neighborhoods, though the military folk made their purchases at shops in the civilian section. The military folk tended to be poorer, but their huts were more sanitary, and they were more likely to seek (or be required to seek) medical attention for plague symptoms.

So when the epidemic started in Kirkee, 1300 of the 1312 military employees received the vaccine, whereas only 45 of the 5640 civilians were vaccinated.

Of those vaccinated from both sides, 0.37% caught plague (0.38% military and 0% civilian), and 0.0007% died from it (0.0008% military and 0% civilian).
Of those not vaccinated, 6.9% caught plague (50% military, 6.8% civilian), and 4.7% died from it (25% military, 4.6% civilian).

Here's a graph of comparative incidence:

Those do seem like very large differences, but the populations of vaccinated vs. unvaccinated were not nearly equivalent, so it may be difficult to compare.

The vaccine, as those before, was made of bacterial culture killed with heat (and a little carbolic acid, aka phenol), and injected into the arm. Typical side effects were localized pain and swelling within 4 hours, going away after 30 hours. Rare individuals had no reaction or had a localized abscess (probably a secondary infection) that was successfully treated. Doesn't sound too bad, especially if it was an effective protection against the plague.

Citation: Lucas, T. C. A note on Haffkine’s antiplague vaccine. Br Med J 1, 928–929 (1907).

010 and 011 - On the Treatment of Acne, Furunculosis, and Sycosis by Therapeutic Inoculations of Staphylococcus Vaccine

Next up is a pair of articles by the same researcher that helped develop the antityphoid vaccine (009), Almroth Edward Wright. Apparently he is quite a character, worth reading about; he predicted the rise of antibiotic-resistant bacteria and was a big advocate of preventative medicine.

In this pair of articles, Wright was attempting to develop a therapeutic (rather than preventative) vaccine against Staphylococcus, a common cause of skin infections such as boils (aka furunculosis) and hair follicle infection (sycosis). The preparation of the vaccine was similar to that of the antityphoid: grow up the bacteria, kill them somehow, and inject this into the patient. Often he used bacteria cultured from the patient’s own infection.

The first paper is a series of case reports, with some in-depth data and graphs(1). The second is another series of descriptive case reports of patients treated with this therapy(2). In each case, it appeared to help the patient’s current problem, but each patient seemed to suffer recurrences, the same or different, pretty often.

One interesting aspect to these reports is the measurement of phagocytic/opsonic index, a measure of how well the treatment increased the immune response. It did seem to increase after the vaccination. Other articles citing these papers give some more details:

Wright and Alexander Fleming developed technique called opsonic/phagocytic index to measure levels of antibodies against pathogen.

"This was considered necessary because by giving too large a dose of vaccine one could lower the antibody level in what was called a 'negative phase' to such an extent that the patient might become worse because he did not have a sufficient margin of resistance to safeguard him against any generalization of the infection....Fleming and his colleagues carried out no less than 16,000 opsonic tests in one year, each of which required half an hour."(3)

"Here again, a lack of adequate controls confounds the analysis. Wright's methods for measuring opsonizing activity were criticized by the renowned biostatistician Karl Pearson who deplored his cavalier indifference to statistical tests of significance: Pearson concludes one paper critical of Wright's data analysis with the exhortation 'Statistics on the table, please!'"(4)

"No adequately controlled clinical trials were undertaken to demonstrate the effectiveness of these forms of therapy."(5)

Sums it up pretty well, I think: not great quality. I hope to learn more about vaccine therapy in future reading though.

References:
1. Wright, A.E. Notes on the treatment of furunculosis, sycosis, and acne by the inoculation of a staphylococcus vaccine,: and generally on the treatment of localised bacterial invasions by therapeutic inoculations of the corresponding bacterial vaccines. The Lancet 159, 874–884 (1902).

2. Wright, A. E. On the Treatment of Acne, Furunculosis, and Sycosis by Therapeutic Inoculations of Staphylococcus Vaccine. Br. Med. J. 1, 1075–1077 (1904).

3. Hare, R. The scientific activities of Alexander Fleming, other than the discovery of penicillin. Med. Hist. 27, 347–372 (1983).

4. Burke, D. S. Vaccine therapy for HIV: A historical review of the treatment of infectious diseases by active specific immunization with microbe-derived antigens. Vaccine 11, 883–891 (1993).

5. Stein, D. S., Timpone, J. G., Gradon, J. D., Kagan, J. M. & Schnittman, S. M. Immune-Based Therapeutics: Scientific Rationale and the Promising Approaches to the Treatment of the Human Immunodeficiency Virus-Infected Individual. Clin. Infect. Dis. 17, 749–771 (1993).

009 - Remarks on the Results which have been Obtained by the Antityphoid Inoculations and on the Methods which have been Employed in the Preparation of the Vaccine

This is another article about the typhoid fever vaccine, more reliable than the previous (008) in my opinion.

Most of the article is a detailed description of how the vaccine may be prepared and tested, which is interesting in itself somewhat; apparently it was made of somewhat-attenuated (that is, less capable of causing severe disease) Salmonella bacteria (the pathogen that causes typhoid fever), grown in culture and then killed by heating or chemical lysis.

It had to be tested for toxicity using guinea pigs—as noted in 008, it was fairly nasty stuff, probably due to all the endotoxin (a component of some bacterial outer membranes that causes immune overreaction and the resulting problems)—and then this toxicity, along with turbidity of the vaccine material, determined the dose to be given to people.

Near the beginning of the article, there are some interesting statistics. A number of British troops in India received vaccination of portions of their soldiers, and the number of cases of typhoid was observed for each. Most of the troops had at least 100 people in each group, usually many more in the unvaccinated group, so it had the potential for a decent comparison. In almost all comparisons, fewer vaccinated than unvaccinated soldiers came down with typhoid, and fewer died from it. The cases that did occur in vaccinated tended to be reported as mild.

Overall, the comparison may not be that impressive (vaccinated: 95 out of each 10000 got sick, 20 in 10000 died; compared to unvaccinated: 250 in 10000 got sick, 34 in 10000 died). But the authors point out that those that received the vaccine tended to be men who were more likely to be susceptible to typhoid in the first place: they were young and had only recently arrived in India, whereas those who were unvaccinated tended to be older and had been exposed to typhoid before, so might have developed natural immunity. In light of that, the numbers seem more impressive. But still, it was an unblinded study with uncertain diagnostic criteria, I think. Consistent with other studies of its kind, though.

Citation: Wright, A. E. & Leishman, W. B. Remarks on the Results which have been Obtained by the Antityphoid Inoculations and on the Methods which have been Employed in the Preparation of the Vaccine.Br Med J 1, 122–129 (1900).

008 - Inoculation with Typhoid Vaccine as a Preventive of Typhoid Fever

Ah, finally the turn of the (20th) century. In this study, the author attempts to test the effectiveness of an anti-typhoid fever vaccine. The subjects were nurses in the typhoid ward at a hospital in Manchester. Fourteen of the 22 subjects received the vaccine, four others refused it, and four had already had typhoid.

Apparently it was a pretty nasty treatment: those that received the vaccine got a day off from work to recover from the headache, general pains, nausea, and vomiting that accompanied vaccination.

Before the intervention, there were about 3-8 cases of typhoid among the nurses each year for almost 5 years, but in the year (or half-year, rather) since vaccination, none of the nurses had caught the disease.

It seems to me like the author should have waited a few more years before publishing anything, to get a good average. To be fair, the half-year post-vaccination period was during typhoid season. But there was no blinding or very effective control group. So I look forward to better studies.

Citation: Marsden, R. W. Inoculation with Typhoid Vaccine as a Preventive of Typhoid Fever. Br Med J 1, 1017–1018 (1900).

007 - Chemical Vaccine against Rabies

Finally we get to something besides smallpox. These researchers in 1893 claim to have extracted material from the central nervous system (CNS) of animals dead from "fixed" rabies (that is, a strain of virus that had been maintained in lab animals) and used this to vaccinate rabbits against the disease. Supposedly when they injected as much as 5 grams of this material in 50 milliliters of liquid under the rabbits' skin, it didn't cause any irritation or side effects at all.

They did some experimental trials with this vaccine too. The first consisted of 14 vaccinated and 6 control rabbits. The former were injected with the vaccine in various quantities between two weeks and one day before being exposed to and infected with "street" rabies (which I assume means the kind you find on the street, that is generally deadly). All of the controls died of the rabies within 21 days, but twelve of the vaccinated rabbits lived at least 5 months (up to the time the study was published, and probably longer) without illness. The two vaccinated that did die of rabies had been given the smallest amounts of vaccine material. But even the one vaccinated right before being infected survived just fine.

This observation probably led to their next trial: vaccinating the rabbits after infecting them with rabies. There were 6 vaccinated and 5 controls in this experiment. The former were vaccinated between 14 hours and 7 days after being infected with "street" rabies. As before, the controls died within 20 days. But four of the six vaccinated did not get sick; the two that did had received either a smaller amount 30 hours after infection, or hadn't been vaccinated until day 7. So vaccinating against rabies works even after being infected with the disease, but it takes more vaccine material and has to be done within a certain incubation period.

These are pretty exciting results. I think if this were a study done today, I wouldn't count it as worth very much (small numbers, only done in animals, probably no blinding), but it seems good for the time, and definitely worth further investigation.

The authors conclude from the lack of side effects and absence of any observable life forms in the extracted vaccine material that the vaccine is chemical in nature, not infective. I don't think viruses had been discovered yet (not until 1898).

It does raise some questions for me though, which would probably be answered if I knew more about rabies. One is, why doesn't the CNS material—taken from animals that died from rabies—cause rabies itself? Maybe the virus dies soon after the animal dies, and then its remnants can be used as a vaccine. How long does the immunity last? Modern pet rabies vaccines require boosters every 2-3 years, but that's still better than nothing. I expect to learn more as I continue to read.

Citation: Tizzoni & Centanni, E. Chemical Vaccine against Rabies. Br Med J 1, 516–517 (1893).

006 - Vaccine and Vaccination:--Observations and Bacteriological Investigations

This was another paper concerned with the success rate of vaccination. The author is concerned with refining the material used to vaccinate so that it is as pure as possible; that is, free from foreign and unnecessary entities. At that time, people were developing the field of microbiology, enough that they had named some of the bacteria they had discovered (some of which retain the same names today).

Unfortunately, they hadn't figured out what the microbe responsible for vaccination was. They thought it might be a kind of coccus, but this wasn't consistently associated with vaccine material. (Makes sense; they hadn't yet realized there were infectious particles even smaller than bacteria.)

However, they did know enough to realize that there could be contamination of the material with pathogens, and the author recommends trying to avoid this.

He reports a study he did with a group of vaccination patients, in which some of them washed the spot to be vaccinated and others did not. Those that washed had lower rates of excessive swelling, only 14% compared to 32% of those that did not wash. Makes sense.

Most importantly, the author speculates that, at least in some cases, ignorant vaccinators mistook the signs of a bacterial infection for the signs that the vaccination was working (i.e. vaccine disease), and so they thought a patient had been effectively vaccinated when actually they were just infected with bacteria. Then when smallpox came along, of course they weren't protected. Hard to know how much of an effect that might have had, but it seems plausible.

Citation: Paquin, P. Vaccine and Vaccination:--Observations and Bacteriological Investigations. Public Health Pap Rep 17, 171–179 (1891).

005 - Failure of vaccine

Vaccination in the 19th century, via inoculation of the arm with some infective material (lymph) from an animal or previous patient, was apparently not always reliable in its effectiveness at protecting one from smallpox. This quote from the introduction of this article seems especially relevant in light of the purpose of this blog:

"The frequent failure of vaccination is now so generally admitted, that statistical proofs are not requisite in order to establish its truth. People look upon it as an equal chance, whether those who have been vaccinated shall be able to resist an attack of the small-pox or not, should they be exposed to it; while some go so far as to surmise, hastily and rashly enough, that vaccination is all but useless....a few, still more inconsiderate, boldly declare themselves in favour of the small-pox itself, as the only and surest guarantee of their safety."

On the other hand, supposedly it was readily apparent to medical men of the time that vaccination had reduced the severity of smallpox outbreaks, despite their unreliability in individual cases. This seems believable.

The author lists a number of hypotheses about why a vaccination might fail to protect, especially faults of the vaccinator such as using lymph from imperfect pocks, inoculating with poor technique, and using lymph from pocks that are too old or too young. He puts particular emphasis on the importance of the appearance and character of the pock from which lymph was taken.

Did these aspects really make much difference? I'm dubious; it seems like as long as some form of vaccine infection was achieved, it should be protective. It doesn't seem like the medical men of the time performed any kind of trials; they just went by their own experiences, a series of anecdotes. I could be wrong, though, and it's not that important now anyway.

Citation: Hingeston, J. A. Failure of vaccine. Assoc Med J 1, 412–414 (1853).

004 - Vaccine Lymph

This is another interesting piece of history, a review of vaccination from the time of Edward Jenner, and a discussion of the best places to get "vaccine lymph," or infectious material used for vaccination against smallpox.

Strangely, multiple people at the time (including a Mr. Badcock) had observed that infecting a cow with smallpox from a human, then using some of the lymph the cow produced from the infection to vaccinate humans, was remarkably effective and mild as a vaccine. From this, the author and his contemporaries conclude that smallpox and cowpox were identical. "This datum is irrefragible."

I'm not so sure, but I don't have a good explanation for this phenomenon. Attenuation, perhaps? I'll be on the lookout for an answer in future readings, and if anyone has any ideas, please comment or email about it.

The author explicitly recognized that vaccination was the lesser of two evils, the ideal situation being that no one would be exposed to either smallpox or the vaccine disease (an ideal fortunately realized in our time), but the latter was obviously preferable to the former.

Citation: Hingeston, J. A. Vaccine Lymph. Assoc Med J 1, 269–273 (1853).

003 - Anomalous Vaccine Eruption

This is a very short report describing a case of smallpox in a child who had just been vaccinated. The child experienced the normal vaccine disease, but then soon after came down with a mild case of smallpox. Another child vaccinated from the diseased material of the first came down with the normal vaccine disease and nothing further, while a third vaccinated from the same source apparently didn't come down with any.

I'm not entirely sure why this is noteworthy. It could, I suppose, show that vaccination does not perfectly protect one against the disease, though it may render it milder. It seems possible that the immunity simply hadn't taken hold yet in the first child, before he caught smallpox (apparently there was a severe outbreak in his locality at the time). Hard to say.

Citation: Hawkins, C. J. Anomalous Vaccine Eruption. Prov Med J Retrosp Med Sci 6, 474 (1843).

002 - Observations on vaccination and small-pox, more especially with reference to the theory of vaccine influence, and the relations subsisting between the cicatrix and the character of the consecutive variola

Another old one, this time from 1841. These old publications are amazingly wordy and flowery! But now we get into some actual interesting data.

As in the previous article, the "vaccination" in this paper refers to inoculating patients with cowpox, also known as the vaccinia virus, pretty much by taking some infected pus from a previous patient and scraping it into a new patient's arm with a needle. This resulted in a mild infection that gave the immune system enough of a recognition of smallpox (variola virus) that it could respond and prevent the latter from being quite as deadly. Remember, smallpox could kill up to 80% of people it infected (I am happy here to be able to use the past tense when referring to it).

The author is a physician at the Smallpox Hospital in London, reporting on an epidemic of smallpox in 1840. Apparently between January 1839 and September 1840, there had only been about 16 cases each month, but then an outbreak started that resulted in almost 16 cases every four days. He gives some statistics about the cases he observed that year.

• 61% of cases had not been vaccinated
• Of these, 45% died
• 38% had been vaccinated previously
• Of these, 7% died
• 1% had had smallpox before

Here's a graph:

Seems like a big difference to me. But it is possible to question how many people of each group got sick enough after being infected to need to go to the hospital, and how many people in the general population of each group got infected at all. All we know is the case:fatality ratios.

Also potentially of interest are the effects on different age groups, of all patients (vaccinated or not):

• 15% were under 5 years old
• 60% of these died
• 14% were 5 to 15 years
• 20% of these died
• 71% were over 15
• 26% of these died
• Overall, 30% of patients died

What the author does report about vaccinated patients is the following: only 9% of those who had been vaccinated were under 16 years, and none were under 5 years (though occasionally a younger vaccinated child did catch it).

Other than that, the rest of the report is arguing that the appearance of the scars on the arms from vaccination  doesn't correlate well with how well-protected an individual is from smallpox. The author gives a series of cases in which nice-looking scars gave little protection and not-nice ones gave good protection.

Overall, it seems like good evidence, for what it's worth, that vaccination helped moderate the severity of smallpox.

Citation: Gregory, G. Observations on vaccination and small-pox, more especially with reference to the theory of vaccine influence, and the relations subsisting between the cicatrix and the character of the consecutive variola. Med Chir Trans 24, 15–29 (1841).

001 - Case of vaccine disease and measles, existing at the same time in the same individual

First, an introduction: there's so much controversy about vaccines, at least in some areas of society (not among scientists, as far as I've seen), so I was inspired to embark on a thorough investigation of the question, so I can promote the truth with more expertise and knowledge. Unfortunately, a Pubmed search for the word "vaccine" turns up more than 220,000 results, and that's just published research; it probably doesn't even cover all the historical, personal, political, and economic questions. But I have to start somewhere, so I'm going to start going through these results.

The first one is the oldest one, from 1823. It is a case report by one S. Gilder, describing his vaccination of a 14-month-old girl.

I wondered what "vaccination" meant in this context, since most of the vaccines we use these days were introduced in the 1950s. But from the context of the article, I figured out that it meant inoculation with cowpox (or in some cases, with smallpox itself) to prevent smallpox. That has been going on for a long time.

So anyway, he inoculated her with cowpox, and then a couple days later she caught measles from her brother. And he observed clear symptoms of measles, at the same time as she was showing clear symptoms of the cowpox inoculation. (Then he applied leeches to treat the measles; this was a while ago.)

The point of the report is to refute an idea present at the time, advocated by one John Hunter, that a person could only suffer from one disease at a time. He derived this theory from several observations of people who, after being vaccinated while ill, seemed to hold off on developing vaccine disease until they got over their illness. But Gilder had just observed two diseases present in the same girl.

It's an interesting piece of medical history, certainly, but I'm not sure it is very helpful for my quest, so I will continue.

Citation: Gilder, S. Case of vaccine disease and measles, existing at the same time in the same individual. Med Chir Trans 12, 186–189 (1823).